TELOMERASE AND THE MECHANISM OF CHROMOSOME END MAINTENANCE

Milica Arneric, Claus M. Azzalin, Klaus Förstemann, Patrick Reichenbach, Peter Sperisen, M. Teresa Teixeira and Joachim Lingner*

Swiss Institute for Experimental Cancer Research (ISREC), CH-1066 Epalinges/s Lausanne, Switzerland

*joachim.lingner@isrec.unil.ch

INTRODUCTION. Telomeres are the nucleoprotein complexes that protect eukaryotic chromosome ends from degradation and fusion. The DNA component of telomeres typically comprises tandem repeats of simple sequences that are rich in guanosine residues in the strand containing the protruding 3’ end. Conventional DNA polymerases cannot synthesize the G-rich 3’ overhang because the parental C-rich strand template is recessed. Telomere length maintenance requires the telomerase enzyme whose RNA moiety (TR) contains the template for tandem telomere repeat synthesis (1). The telomerase reverse transcriptase subunit (TERT) bears structural similarity with reverse transcriptases from retroelements and provides the active site (1).

THE TELOMERASE REACTION CYCLE. During processive telomere extension, telomerase repeatedly uses the same small region within its RNA moiety as a template for DNA synthesis. Following each telomeric repeat addition, the RNA template and the telomeric substrate reset their relative position in the active site of TERT. The translocation reaction requires disruption of base-pairs in the DNA/RNA hybrid, repositioning of the RNA template relative to the active site and reformation of DNA/RNA base-pair interactions at the other end of the template. We have partially purified yeast telomerase and used dimethyl sulfate modification to assess structural changes of the RNA template during the reaction cycle. The A’s and C’s in the template were accessible in the absence of the DNA oligonucleotide substrate and reacted with DMS. As expected, the template bases became partially protected upon substrate binding but the base-pairing interactions encompassed only the most seven 3’ proximal nucleotides of the substrate. This number of base-pairs was maintained upon substrate elongation at the 3’ end. This therefore provides direct evidence that telomerase limits the number of base-pairs between substrate and template during the reaction cycle. Thus, the activation energy for translocation may be lower than what would be anticipated for a fully paired telomeric DNA-telomerase RNA hybrid.

TELOMERASE RECRUITMENT AND TELOMERE CAPPING. The TERT-TR telomerase core is sufficient for catalytic telomerase activity in vitro but additional factors are required for telomerase activation at chromosomal ends. In the yeast Saccharomyces cerevisiae, the Ever Shorter Telomeres 1 (EST1) gene product, recruits telomerase to the 3’ end of telomeres (2) allowing telomere elongation in S-phase. It is unclear if similar mechanisms of telomerase regulation have been conserved in vertebrates during evolution. By iterative profile searches we and others identified a human homolog of yeast Est1p (hEST1A) that associates with at least 70% of active telomerase in HeLa cells. Over-expression of hEST1A induced rapid growth arrest and apoptotic-like cell death in several human cancer cell lines as well as in hTERT-negative primary fibroblasts. Analysis of metaphase chromosomes uncovered a time-dependent accumulation of chromosomal end-to-end fusions leading to anaphase bridges without perturbing the length of the telomeric tract. Thus, over-expression of hEST1A triggers chromosome uncapping. To further elucidate the function of hEST1A we have analyzed the phenotype of cells in which endogenous hEST1A was downregulated by siRNA. hEST1A depletion induced rapid loss of telomeric DNA in HeLa cells, which was accompanied by cell growth arrest and severe morphological changes. The telomeric-loss rate was much faster than what was expected from an inactivation of a functional telomerase pathway. Thus it appears that hEST1A plays a critical role in protecting chromosome ends from nucleolytic degradation or telomere rapid deletion events. A protective role was recently also described for an EST1-homolog in Candida albicans. In summary the data may indicate that EST1-like proteins have dual roles in telomerase recruitment and telomere protection.

TELOMERE LENGTH HOMEOSTASIS. In cells that express telomerase, the length of the duplex telomeric repeat array is kept within a species- and cell type-specific narrow range. Telomere length homeostasis is the result of a balance between telomere shortening and telomere lengthening activities. Previous work has indicated that telomerase activity is regulated in cis at individual telomeres through the number of double-strand telomere binding proteins (3). At least two not mutually exclusive models could provide a mechanistic basis for the protein-counting model of telomere length control and the increased activity of telomerase on shorter telomeres. First, the elongation efficiency of telomerase, i.e. the number of nucleotides added to an individual telomere per elongation event, could be regulated as a function of telomere length. Thus, all telomeres would be available for telomerase-mediated extension but the telomere structure would regulate the catalytic activity of telomerase in a length-dependent manner, perhaps by influencing the processivity or turnover of the enzyme. Second, the productive association of telomerase with telomere 3’ ends could be regulated by length-dependent changes in telomeric chromatin structure. A long telomere would have a lower probability to be in a telomerase-extendible state than a short telomere. In this model, a telomere could shorten for several rounds of DNA replication without being elongated by telomerase before its chromatin structure would switch and become competent for telomerase-mediated elongation. Thus, the association rather than the activity of telomerase would be the regulated element. To distinguish between these models, we devised a system to measure elongation of single telomeres in vivo in S. cerevisiae at nucleotide resolution. We find that telomerase does not act on every telomere in every cell cycle. Instead, it exhibits an increasing preference for telomeres as their lengths decline. The number of nucleotides that are added to a telomere in a single cell cycle varies between a few to more than 100 nucleotides and is independent of telomere length. Deletion of the telomeric protein Rif1p gives rise to longer telomeres by increasing the frequency of elongation events. Thus, by taking a molecular snapshot of the result of a single round of telomere replication, we demonstrate that telomere length homeostasis is achieved via a switch between telomerase extendible and non-extendible states.

ACKNOWLEDGMENT. We thank Lea Harrington (Toronto) and Julie Cooper (London) for communication of results prior to publication and Titia de Lange (New York), Lea Harrington and Vicki Lundblad (Houston) for sharing material.

REFERENCES

1. Kelleher, C., Teixeira, M. T., Forstemann, K., and Lingner, J. (2002). Trends Biochem. Sci. 27, 572-579
2. Evans, S. K., and Lundblad, V. (1999) Science 286, 117-120.
3. Marcand, S., Gilson, E., and Shore, D. (1997) Science 275, 986-990.